Does anyone know of any quantitative studies that look at the effect of autoplaying videos on the user experience of a website (i.e. a page loads and—BOOM—a video + audio starts playing)? My UX senses tell me this is a bad experience for many users, and I know that many qualitative arguments have been made against the practice (like in this UX.StackExchange thread). What I need, though, is hard evidence of how autoplaying videos degrade user experience (or improve it—I'm willing to admit that I may be totally wrong).
So, anyone know of studies that offer such evidence? Peer-reviewed studies would be ideal, but any studies that have actual data to back up their assertions would work.